Abstract
The Optimal Fuel Shuffling System (OFSS) is developed for optimal design of PWR fuel loading pattern. In this paper, an optimal loading pattern is defined that the local power peaking factor is lower than predetermined value during one cycle and the effective multiplication factor is maximized in order to extract maximum energy. OFSS is a hybrid system that a rule based system, a fuzzy logic, and an artificial neural network are connected each other. The rule based system classifies loading patterns into two classes using several heuristic rules and a fuzzy rule. A fuzzy rule is introduced to achieve more effective and fast searching. Its membership function is automatically updated in accordance with the prediction results. The artificial neural network predicts core parameters for the patterns generated from the rule based system. The back-propagation network is used for fast prediction of core parameters. The artificial neural network and the fuzzy logic can be used as the tool for improvement of existing algorithm's capabilities. OFSS was demonstrated and validated for cycle 1 of Kori unit 1 PWR. (author)
Kim, Han Gon;
Chang, Soon Heung;
Lee, Byung
[1]
- Department of Nuclear Engineering, Korea Advanced Institute of Science and Technology, Yusong-gu, Taejon (Korea, Republic of)
Citation Formats
Kim, Han Gon, Chang, Soon Heung, and Lee, Byung.
A study on the optimal fuel loading pattern design in pressurized water reactor using the artificial neural network and the fuzzy rule based system.
IAEA: N. p.,
2004.
Web.
Kim, Han Gon, Chang, Soon Heung, & Lee, Byung.
A study on the optimal fuel loading pattern design in pressurized water reactor using the artificial neural network and the fuzzy rule based system.
IAEA.
Kim, Han Gon, Chang, Soon Heung, and Lee, Byung.
2004.
"A study on the optimal fuel loading pattern design in pressurized water reactor using the artificial neural network and the fuzzy rule based system."
IAEA.
@misc{etde_20529203,
title = {A study on the optimal fuel loading pattern design in pressurized water reactor using the artificial neural network and the fuzzy rule based system}
author = {Kim, Han Gon, Chang, Soon Heung, and Lee, Byung}
abstractNote = {The Optimal Fuel Shuffling System (OFSS) is developed for optimal design of PWR fuel loading pattern. In this paper, an optimal loading pattern is defined that the local power peaking factor is lower than predetermined value during one cycle and the effective multiplication factor is maximized in order to extract maximum energy. OFSS is a hybrid system that a rule based system, a fuzzy logic, and an artificial neural network are connected each other. The rule based system classifies loading patterns into two classes using several heuristic rules and a fuzzy rule. A fuzzy rule is introduced to achieve more effective and fast searching. Its membership function is automatically updated in accordance with the prediction results. The artificial neural network predicts core parameters for the patterns generated from the rule based system. The back-propagation network is used for fast prediction of core parameters. The artificial neural network and the fuzzy logic can be used as the tool for improvement of existing algorithm's capabilities. OFSS was demonstrated and validated for cycle 1 of Kori unit 1 PWR. (author)}
place = {IAEA}
year = {2004}
month = {Jul}
}
title = {A study on the optimal fuel loading pattern design in pressurized water reactor using the artificial neural network and the fuzzy rule based system}
author = {Kim, Han Gon, Chang, Soon Heung, and Lee, Byung}
abstractNote = {The Optimal Fuel Shuffling System (OFSS) is developed for optimal design of PWR fuel loading pattern. In this paper, an optimal loading pattern is defined that the local power peaking factor is lower than predetermined value during one cycle and the effective multiplication factor is maximized in order to extract maximum energy. OFSS is a hybrid system that a rule based system, a fuzzy logic, and an artificial neural network are connected each other. The rule based system classifies loading patterns into two classes using several heuristic rules and a fuzzy rule. A fuzzy rule is introduced to achieve more effective and fast searching. Its membership function is automatically updated in accordance with the prediction results. The artificial neural network predicts core parameters for the patterns generated from the rule based system. The back-propagation network is used for fast prediction of core parameters. The artificial neural network and the fuzzy logic can be used as the tool for improvement of existing algorithm's capabilities. OFSS was demonstrated and validated for cycle 1 of Kori unit 1 PWR. (author)}
place = {IAEA}
year = {2004}
month = {Jul}
}